Hello πŸ‘‹

Creating your AI voice with Deepsync requires no effort from your side. You don't have to do any editing or read a long script, all that is in the past. With your previously recorded audio files, we can create your high-quality AI voice. We currently require anywhere from 30 minutes to start creating a voice. For the best quality, about 5 to 10 hours of audio is required. For a much faster onboarding, you can simply share your RSS feed with us to get the process started. Deepsync currently supports all accents in the English language.

Uploading your audio files

To upload your audio files directly, please sign up on our Hompage and we'll set up your account where you can securely upload your audio data.

Using RSS Feed

To upload your audio files using the RSS feed, please sign up on Homepage and we'll set up your account and you can simply paste your RSS feed URL on your Dashboard. If you don't know your RSS feed URL, you can find that by following this link. We may ask for verification in order to verify if the RSS feed belongs to you.

Please note that we require a short recorded consent from your side to perform a vocal identification before we can begin creating your voice. Once you provide the same, it would take almost a week for your voice to get ready.

To create a perfect AI voice, please ensure that your audio data meets these basic qualifications for the best experience:

  1. Your audios contain your own voice and do not contain other voices such as a podcast guest. Although this part is optional, we recommend having a solo voice for the AI to learn faster and better.

  2. If your audio files contain a fixed Intro/Outro music, that should be fine as long as the background music/noise does not occur in random places in the audio since it'll be harder to find and clean that up.

  3. Your audios are of consistent quality i.e. they are recorded in the same surrounding. This ensures that your AI voice sounds consistent throughout the produced Script.

Here's a sample of an ideal audio recording:

... and that's it! Once we verify your audio data, we will begin the process and you'll never have to record anything manually again (for the most part)!

Next up, read about our Dashboard ↗️

Did this answer your question?